EFFICIENT CLUSTERING-TECHNIQUES FOR CHANNEL EQUALIZATION IN HOSTILE ENVIRONMENTS

In this paper the equalization problem is treated as a classification task. No specific (linear or nonlinear) model is required for the chan nel or for the interference and the noise. Training is achieved via a supervised learning scheme. Adopting Mahalanobis distance as an approp riate distance metric, decisions are made on the basis of minimum dist ance path. The proposed equalizer operates on sequence mode and implem ents the Viterbi searching Algorithm. The robust performance of the eq ualizer is demonstrated for a hostile environment in the presence of C CI and nonlinearities, and it is compared against the performance of t he MLSE and a symbol by symbol RBF equalizer. Suboptimal techniques wi th reduced complexity are discussed. (C) 1997 Elsevier Science B.V.